High temperature causes breakdown of S haplotype-dependent stigmatic self-incompatibility in self-incompatible Arabidopsis thaliana

High temperature disrupts the targeting of SRK to the plasma membrane, resulting in breakdown of the stigmatic self-incompatibility response in self-incompatible Arabidopsis transformants.

Abstract

Commercial seeds of Brassicaceae vegetable crops are mostly F₁ hybrids, the production of which depends on self-incompatibility during pollination. Self-incompatibility is known to be weakened by exposure to elevated temperatures, which may compromise future breeding and seed production. In the Brassicaceae, self-incompatibility is controlled by two genes, SRK and SCR, which function as female and male determinants of recognition specificity, respectively. However, the molecular mechanisms underlying the breakdown of self-incompatibility under high temperature are poorly understood. In this study, we examined the self-incompatibility phenotypes of self-incompatible Arabidopsis thaliana SRK-SCR transformants under normal (23 °C) and elevated (29 °C) temperatures. Exposure to elevated temperature caused defects in the stigmatic, but not the pollen, self-incompatibility response. In addition, differences in the response to elevated temperature were observed among different S haplotypes. Subcellular localization revealed that high temperature disrupted the targeting of SRK to the plasma membrane. SRK localization in plants transformed with different S haplotypes corresponded to their self-incompatibility phenotypes, further indicating that defects in SRK localization were responsible for the breakdown in the self-incompatibility response at high temperature. Our results provide new insights into the causes of instability in self-incompatibility phenotypes.

Introduction

Plant growth and development are affected by variability in environmental conditions. Both higher average and peak temperatures caused by climate change have the potential to affect the performance of crop plants worldwide. The Brassicaceae family contains a number of agriculturally and horticulturally significant crops. Commercial cultivation of Brassicaceae vegetables generally employs F₁ hybrids, which show superior crop performance relative to their parents (Lippman and Zamir, 2007; Fujimoto et al., 2018). F₁ hybrid seeds of Brassica vegetables are produced by mutual crossing of two inbred self-incompatible parent lines, and also by crossing a cytoplasmic male-sterile line as a female parent with a male-fertile parent line. However, the self-incompatibility is known to be erratic, as external factors such as high temperature (Nasrallah and Wallance, 1968; Richards and Thurling, 1973; Visser, 1977; Hirosaki and Niikura, 2008) as well as internal factors can cause breakdown of the self-incompatibility response (Carter and McNeilly, 1975; Ockendon, 1978; Hirosaki and Niikura, 2008). As a result, F₁ hybrid seed populations obtained by using self-incompatibility often contain some self-fertilized and sibling-fertilized seeds, and this is an important issue for the production of F₁ hybrid seed.

The Brassicaceae family has a sporophytic self-incompatibility mechanism that is governed by haplotypes of the S locus, which contains two tightly linked polymorphic genes: S-locus receptor kinase (SRK) and S-locus cysteine-rich protein/S-locus protein 11 (SCR /SP11, hereafter referred to as SCR). SRK encodes a plasma membrane-localized receptor kinase that is expressed in stigma papilla cells (Stein et al., 1991; Takasaki et al., 2000), and SCR encodes a pollen coat-localized peptide ligand for SRK (Schopfer et al., 1999; Takayama et al., 2000). SRK can only interact with SCR of the same S haplotype (Kachroo et al., 2001; Takayama et al., 2001). Allele-specific SRK–SCR interactions trigger the self-incompatibility response, causing inhibition of pollen germination and of pollen tube penetration into the stigma epidermal cell wall.

Several SRK+SCR gene pairs, including the S_b haplotype in Arabidopsis lyrata, which is a closely related self-incompatible species to A. thaliana, have been previously shown to confer an intense self-incompatibility phenotype in the A. thaliana C24 ecotype (Nasrallah et al., 2004; Boggs et al., 2009a, b). The AlSRK+AlSCR A. thaliana transformant provides a very useful model system for analysis of the self-incompatibility mechanism in the Brassicaceae, which is difficult to examine using Brassica crops. These transformants have provided new insights into a variety of molecular mechanisms, such as SRK–SCR interactions (Boggs et al., 2009a), SRK biogenesis (Yamamoto et al., 2014), and self-incompatibility signaling (Kitashiba et al., 2011; Yamamoto and Nasrallah, 2013; Iwano et al., 2015).

Breakdown of self-incompatibility in Brassicaceae as a result of high temperatures has been known for more than 50 years. Nasrallah and Wallance (1968) examined the effects of temperature on the degree of self-incompatibility by cultivating cabbage plants in growth chambers set at 15, 21, and 27 °C. They found that a high-temperature (27 °C) pre-treatment for 6 d before pollination caused an increase in the number of seeds after selfing, and that the differences of sensitivity of self-incompatibility to high temperature were likely to be determined by the genetic background of the plants. Brussels sprouts grown in glasshouses controlled at 23/14 °C and 26/14 °C (day/night, daylength 11 h) have been shown to produce more seeds than those grown at 17/14 °C and 20/14 °C (Visser, 1977). Both studies reported that plants showing a high degree of the self-incompatibility response were not affected by the high-temperature treatment (Nasrallah and Wallance, 1968; Visser, 1977). Richards and Thurling (1973) found that Brassica rapa plants treated at 30 °C in a temperature-controlled growth cabinet showed lower self-incompatibility levels than those at 25 °C. However, since S genes had not been identified in the 1960s and 1970s, the roles of the SRK and SCR proteins in self-incompatibility instability under high temperature conditions and their molecular mechanisms remained unclear.

In this study, the molecular mechanisms behind the breakdown of self-incompatibility resulting from high-temperature treatment was investigated using self-incompatible A. thaliana plants transformed with SRK and SCR gene pairs derived from A. lyrata. Self-incompatible transgenic A. thaliana can be cultivated in growth chambers under strictly controlled temperatures, thus serving as a suitable model for the analysis of molecular mechanisms of high temperature effects. We examined the self-incompatibility of self-incompatible A. thaliana transformants under normal (23 °C) and elevated temperature (29 °C) by pollination assays and seed production, and monitored the subcellular localization of SRK proteins by biochemical analysis. We found that high temperature perturbed the stigmatic self-incompatibility response. Our results indicated that defects in plasma-membrane localization of SRK are responsible for the breakdown in self-incompatibility at high temperature.

Materials and methods

Plant material and growth conditions

Arabidopsis thaliana C24 plants (obtained from the Arabidopsis Biological Resource Center, https://abrc.osu.edu/) were used in this study. All plants were grown at a constant temperature of 23 °C under long-days (100 µmol photons m^–2 s^–1, 16/8 h light/dark). The relative humidity was not controlled. To avoid plant age-related effects on seed fertility and self-incompatibility phenotypes when counting seed numbers, at least eight siliques were collected between 1–3 weeks after flowering and seeds per silique were counted. For the high-temperature treatment, plants were grown at 23 °C until bolting and then transferred to growth chambers set at a constant temperature of 29 °C under long days as detailed above.

Plasmid construction and production of transgenic plants

AlSRK _b-FLAG+AlSCR_bA. thaliana transformants have been described previously by Yamamoto et al. (2014). AlSRK_b+AlSCR_b, AlSRK_a+AlSCR_a, AlSRK₃₉+AlSCR₃₉, and AlSRK₃₉-FLAG+AlSCR₃₉A. thaliana transformants were developed as follows. Arabidopsis lyrata plants carrying the S_a and S₃₉ haplotypes were grown from seeds (kindly provided by Professor Barbara Mable, University of Glasgow, UK) and were identified by nucleotide sequence analysis of PCR products amplified with the 13F1 (ccgacggtaaccttgtcatcctc) and SLGR (atcgacataaagatcttgacc) primers (Schierup et al., 2001) using a GenomeLab DTCS Quick Start Kit (SCIEX, Framingham, MA) and a Beckman Coulter CEQ2000XL DNA sequencer (SCIEX). A 360-bp fragment corresponding to the AtS1 (At3g12000) promoter, a stigma papilla cell-specific promoter (Yamamoto et al., 2014), was used to control expression of the AlSRK_a and AlSRK₃₉ genes. The AtS1 promoter and a ~350-bp DNA fragment corresponding to the A. lyrata SRK_b terminator were introduced into the EcoRI and KpnI sites and the BamHI and XbaI sites of pCambia1300, respectively, and the resulting plasmid was designated as pMYC290. The AlSRK_b, AlSRK_a, and AlSRK₃₉ coding sequence fragments were amplified by PCR using the primers AlSRKb-F and AlSRKb-R for AlSRK_b, primers AlSRKa-F and AlSRKa-R for AlSRK_a, and primers AlSRK39-F and AlSRK39-R for AlSRK39 (see Supplementary Table S1 at JXB online). The amplified AlSRK_b coding sequence fragment was introduced into the KpnI site of pMYC290 by InFusion cloning (Takara Bio), followed by introduction of the AlSCR_b genomic fragment containing ~580 bp and ~220 bp regions corresponding to the AlSCR_b promoter and terminator, respectively, into the XbaI site. A 2680-bp AlSCR_a genomic fragment and a 2952-bp AlSCR₃₉ genomic fragment, which contained 600 bp and 300 bp of the promoter and terminator, respectively, were amplified by PCR with the primers AlSCRa-F and AlSCRa-R for AlSCR_a and primers AlSCR39-F and AlSCR39-R for AlSCR₃₉ (Supplementary Table S1). The AlSRK_a fragment was introduced into the KpnI site of pMYC290 by InFusion cloning (Takara Bio), followed by introduction of the AlSCR_a fragment into the XbaI site. The AlSRK₃₉ and AlSCR₃₉ fragments were similarly introduced into the KpnI and PstI sites, respectively, of pMYC290 by InFusion cloning (Takara Bio). For construction of AlSRK₃₉-FLAG+AlSCR₃₉, the AtS1 promoter-StuI site-3×FLAG-AlSRK_b-terminator DNA cassette was produced by the recombinant PCR method (Pont-Kingdon, 1994). Genomic DNA prepared from the AlSRK_b-FLAG+AlSCR_bA. thaliana transformant was used as the template, using the primers AtS1 StuI-F and AtS1 StuI-R for the AtS1 promoter and the primers AlSRKb term StuI-F and AlSRKb term StuI-R for the AlSRK_b terminator (Supplementary Table S1) in the first-round PCR. Next, the AtS1 promoter-StuI site-3×FLAG-AlSRK_b-terminator fragment was amplified using the first-round PCR products as templates with the AtS1 StuI-F and AlSRKb term StuI-R primers (Supplementary Table S1). The resulting AtS1 promoter-StuI site-3×FLAG-AlSRK_b-terminator fragment was introduced into the EcoRI site of pCambia1300, and the construct was designated as pMYC190. The AlSCR_b promoter and AlSCR_b terminator were amplified by PCR using the primers AlSCRb pro-F and AlSCRb pro-R and the primers AlSCRb term-F and AlSCRb term-R, respectively (Supplementary Table S1). The amplified fragments were introduced into the SacI and KpnI sites and the PstI and HindIII sites of pMYC190, respectively, and the resulting construct was designated as pMYC192. The AlSRK₃₉ and AlSCR₃₉ genomic fragments were amplified by PCR using the primers AlSRK39-FLAG-F and AlSRK39-FLAG-R and the primers AlSCR39-F2 and AlSCR39-R2, respectively (Supplementary Table S1). The AlSCR₃₉ fragment was inserted into the KpnI and SalI sites of pMYC192, followed by introduction of the AlSRK₃₉ fragment into the StuI site by InFusion cloning (Takara Bio). The resulting plasmids were used for transformation of A. thaliana plants by the floral dip method (Clough and Bent, 1998) using Agrobacterium tumefaciens strain GV3101.

Pollination assays

We carried out the pollination assays in the morning (usually between 2–3.5 h after the beginning of the light period). Emasculated flowers at flower stage 13 (Smyth et al., 1990), which is 1 d before anthesis in A. thaliana, were collected and placed on a 0.5% agar plate, and stigma were manually pollinated under a stereomicroscope with pollen grains from mature flowers. They were then left for 2 h at 23°C under continuous light (relative humidity was not controlled) prior to fixation in an ethanol:acetic acid (3:1) solution for 10 min at 55 °C. The fixed stigmas were treated with 8 M NaOH for 15 min at room temperature, then washed with water, and stained with decolorized Aniline Blue. Stained stigmas were observed using fluorescence microscopy as previously described (Kho and Bear, 1968). Each pollination assay was performed with at least three stigmas, and was repeated on at least two different dates for each transformant line. Pollen tubes were counted, and self-incompatibility was classified according to the number of tubes per pollinated stigma as follows: incompatible, <10; partially incompatible, 10–29; compatible, >30. Images of pollinated stigmas were captured using an Axioskop microscope with an AxioCam ERc 5s camera (Carl Zeiss) or a BX51 microscope with a DP70 camera (Olympus). For the high-temperature treatment before pollination, A. thaliana plants were transferred from the growth chambers at 23 °C to chambers at 29 °C, and left for either 6 h or 16 h under continuous light (100 µmol photons m^–2 s^–1) before the pollination assay. For the high-temperature treatment after pollination, manual pollination was carried out as described above using A. thaliana plants grown at 23 °C under long days, and the pollinated pistils on the agar plates were placed in the growth chamber set at 29 °C for 16 h. The relative humidity was not controlled.

Protein electrophoresis

Ten flower buds at flower stage 13 were collected from each of the untreated plants (23 °C) and plants exposed to 29 °C for 16 h and homogenized in extraction buffer [100 mM Tris-HCl, pH 8.0, 2% (w/v) SDS, 5.7 mM β-mercaptoethanol, and 1 mM phenylmethylsulfonyl fluoride]. The extracted proteins were precipitated using the trichloroacetic acid/acetone method and dissolved in SDS-PAGE sample buffer (Laemmli, 1970). Some of the protein samples were treated with endoglycosidase H (New England BioLabs) at 37 °C for 1 h. The proteins were separated on a 7.5% SDS-PAGE gel and transferred to Immobilon-P membranes (Millipore) as previously described (Towbin et al., 1979). For detection, primary anti-FLAG antibody (Sigma-Aldrich) was used at 1:1000 dilution, and secondary HRP-labeled anti-mouse IgG (GE Healthcare) was used at 1:3000 dilution. The protein membranes were subsequently stained with Coomassie Brilliant Blue R250 and detected proteins were quantified using the ImageJ software (https://imagej.nih.gov/ij/).

Results and discussion

Breakdown of the self-incompatibility response in A. thaliana AlSRK_b+AlSCR_b stigmas upon exposure to elevated temperature

To examine the effect of high temperature on self-incompatibility in self-incompatible plants, A. thaliana C24 plants expressing a C-terminal FLAG-tagged AlSRKb and AlSCRb (hereafter AlSRK_b-FLAG+AlSCR_b), which had previously been found to exhibit an intense self-incompatibility response (Yamamoto et al., 2014), were subjected to high temperature (29 °C) for 6 h or 16 h prior to pollination. Self-incompatibility was generally assessed following manual pollination of stage-13 flower buds (Smyth et al., 1990) as the number of pollen tubes per pollinated stigma. No pollen germination or pollen tube growth was observed in AlSRK_b-FLAG+AlSCR_bA. thaliana transformants that had not been exposed to elevated temperatures (Fig. 1A). Stigmas treated at 29 °C for 6 h showed some self-pollen germination but no pollen tube growth. In contrast, treatment at 29 °C for 16 h resulted in self-pollen germination and pollen tube penetration into the stigma tissues.

Fig. 1.

Self-incompatibility phenotypes of AlSRK_b+AlSCR_bArabidopsis thaliana transformants at high temperature. Plants were grown at 23 °C. For the high-temperature treatment, plants were transferred to 29 °C for either 6 h or 16 h before being hand-pollinated. Schematic diagrams above the images indicate the procedures used for the pollination assays. Aniline Blue staining was used to monitor pollen germination and pollen tube growth. (A) Images of self-pollinated AlSRK_b-FLAG+AlSCR_b [AlSRK_b-FLAG +AlSCR_b] stigmas. Numerous self-pollen tubes were observed in AlSRK_b-FLAG+AlSCR_b stigmas treated at 29 °C for 16 h. (B) Pistils of wild-type (WT) and AlSRK_b-FLAG+AlSCR_b transformants grown at 23 °C were self-pollinated and incubated at 29 °C for 16 h. Self-pollen tubes were not observed in AlSRK_b-FLAG+AlSCR_b stigmas. (C) Upper images show stigmas of AlSRK_b-FLAG+AlSCR_b transformants treated at high temperature and pollinated with pollen from AlSRK_b-FLAG+AlSCR_b transformants (left) or WT plants (right) grown under normal conditions (i.e. 23 °C). Lower panels show the stigmas of AlSRK_b-FLAG+AlSCR_b transformants (left) and the WT (right) grown under normal conditions and pollinated with pollen from AlSRK_b-FLAG+AlSCR_b transformants treated at high temperature. Numerous pollen tubes elongated in the stigmas of AlSRK_b-FLAG+AlSCR_b transformants treated at 29 °C. By contrast, pollen from the AlSRK_b-FLAG+AlSCR_b transformants treated at 29 °C exhibited an intense incompatible response on the stigmas of AlSRK_b-FLAG+AlSCR_b transformants. Scale bars are 100 µm. (This figure is available in colour at JXB online.)

To investigate the effect of high temperature after pollination on self-incompatibility, stigmas of AlSRK_b-FLAG+AlSCR_bA. thaliana transformants grown at 23 °C were self-pollinated, and the pollinated pistils were incubated for 16 h at 29 °C. No self-pollen germination or elongation was observed on the stigmas (Fig. 1B). Exposure to high temperature was not responsible for the lack of pollen germination since wild-type (WT) pollen grains germinated and pollen tubes elongated in the stigmas of WT plants (Fig. 1B). These experiments indicated that high-temperature treatment for 16 h before, but not after, pollination caused defects in self-incompatibility.

We also investigated whether high temperature affected the self-incompatibility response in the stigmas, the pollen, or both stigmas and pollen. When the stigmas of AlSRK_b-FLAG+AlSCR_bA. thaliana transformants treated at 29 °C were pollinated with pollen from untreated transformants, numerous pollen tubes were observed in the pistils (Fig. 1C), as was also observed with pollen from WT plants. By contrast, pollen from AlSRK_b-FLAG+AlSCR_bA. thaliana transformants treated at 29 °C did not germinate on the stigmas of transformants grown at 23 °C. Germination and pollen tube growth of pollen from AlSRK_b-FLAG+AlSCR_bA. thaliana transformants treated at 29 °C were observed on untransformed WT stigmas, indicating that the high-temperature treatment did not disrupt the viability of AlSCRb-bearing pollen. These results indicated that the high-temperature treatment did not affect self-incompatibility in the pollen.

Intense self-incompatibility response in AlSRK₃₉+AlSCR₃₉ A. thaliana transformants at 29 °C

Next, we examined differences in the effects of high temperature in self-incompatible A. thaliana plants transformed with different A. lyrata S haplotypes. Binary vectors harboring AtS1_pro::AlSRK_a+AlSCR_apro::AlSCR_a or AtS1_pro::AlSRK₃₉+AlSCR_39pro::AlSCR₃₉ were constructed, in which AlSRK was placed under the control of the stigma papilla cell-specific AtS1 promoter (Dwyer et al., 1994) and AlSCR under the control of its own promoter. The constructs, as well as an AtS1_pro::AlSRK_b+AlSCR_bpro::AlSCR_b construct, were introduced into the A. thaliana C24 ecotype. To confirm that AlSRK_a+AlSCR_a and AlSRK₃₉+AlSCR₃₉ conferred self-incompatibility activity, A. thaliana transformed stigmas were manually self-pollinated, and pollen germination and pollen tube growth were monitored by fluorescence microscopy. Four of ten AlSRK_a+AlSCR_a transformants, 14 of 15 AlSRK₃₉+AlSCR₃₉ transformants, and 9 of 15 AtS1_pro::AlSRK_b+AlSCR_bpro::AlSCR_b transformants showed robust self-incompatibility under normal temperature conditions (Table 1 and Fig. 2A).

Table 1.

Pollination phenotypes of transgenic Arabidopsis thaliana expressing A. lyrata S genes at 23 °C

Transgenes	Total no. of transformants	No. of self-pollen tubes observed per stigma*
		<10	10–29	>30
AlSRK b+AlSCRb	15	9	0	6
AlSRK a+AlSCRa	10	4	0	6
AlSRK 39+AlSCR39	15	14	0	1
AlSRK 39 -FLAG +AlSCR39	11	6	0	5

* Number of pollen tubes observed in stigma 2 h after pollination with self-pollen: <10, incompatible; 10–29, partially incompatible; >30, compatible.

Fig. 2.

The effects of high temperature on the self-incompatibility response in Arabidopsis thaliana plants expressing different A. lyrata S haplotypes. (A) The self-incompatibility phenotypes of AlSRK_b+AlSCR_b (left), AlSRK_a+AlSCR_a (center), and AlSRK₃₉+AlSCR₃₉ (right) transformants at normal temperature (23 °C, upper panels) or treated at 29 °C for 16 h (lower panels) before pollination. A schematic diagram above the images indicates the procedure used for the pollination assay. Aniline Blue staining was used to monitor pollen germination and pollen tube growth. Transformants expressing AlSRK_b+AlSCR_b and AlSRK_a+AlSCR_a exhibited complete breakdown of the self-incompatibility response after exposure to elevated temperature prior to pollination, whereas transformants expressing AlSRK₃₉+AlSCR₃₉ retained the self-incompatibility response. Scale bar is 100 µm. (B) Box-plots of seed number per silique for plants grown at 23 °C or 29 °C. Data are 10 siliques for AlSRK_b+AlSCR_b transformant #1 grown at 23 °C, and from 8 siliques for the other transformants. Different letters indicate significant differences between means as determined using the Tukey–Kramer method (P<0.05). (This figure is available in colour at JXB online.)

The AlSRK+AlSCR self-incompatible transformants were exposed to high temperature (29 °C) for 16 h, then they were self-pollinated, and self-pollen germination and tube elongation were examined. The heat-treated stigmas of three of four self-incompatible AlSRK_a+AlSCR_a transformants (transformants #2, #5, and #7) allowed numerous self-pollen tubes to penetrate into the stigmas (Fig. 2A). This phenotype was indistinguishable from that of the stigmas of two self-incompatible AlSRK_b+AlSCR_bA. thaliana transformants (transformants #1 and #2) treated at 29 °C. These observations indicated that the high-temperature treatment also caused defects in the self-incompatibility of A. thaliana expressing AlSRK_a+AlSCR_a. By contrast, stigmas of three of the tested self-incompatible AlSRK₃₉+AlSCR₃₉ transformants (transformants #1, #2, and #4) retained intense self-incompatibility when exposed to high temperature (Fig. 2A). These results suggested that stigmas expressing different A. lyrata S haplotypes differed in their sensitivity to the high-temperature treatment.

Seed production phenotypes of the self-incompatible AlSRK+AlSCR A. thaliana transformants at 23 °C and 29 °C

Seed production was examined under normal (23 °C) and high (29 °C) temperature conditions in the AlSRK+AlSCR transformants that exhibited a self-incompatibility response in the pollination assays (Table 1). Wild-type plants grown at 23 °C contained 46.4±4.4 seeds per silique whereas the AlSRK_b+AlSCR_b and AlSRK₃₉+AlSCR₃₉A. thaliana transformants contained significantly fewer (Fig. 2B). Although the AlSRK_a+AlSCR_aA. thaliana transformants exhibited self-incompatibility, the number of seeds per silique produced under normal conditions in AlSRK_a+AlSCR_a transformants #5 and #7 did not differ significantly from that of the WT plants, and AlSRK_a+AlSCR_a transformant #2 produced more seeds than the AlSRK_b+AlSCR_b and AlSRK₃₉+AlSCR₃₉ transformants (Fig. 2B). The S_a haplotype is recessive to S_b and S₃₉ (Goubet et al., 2012), and one possible explanation for these results is that the self-incompatibility response determined by the S_a haplotype may be weaker than that determined by the S_b and S₃₉ haplotypes. This suggestion is supported by a previous study that showed that recessive S haplotypes produce weaker self-incompatibility phenotypes than dominant S haplotypes in Brassica (Nasrallah and Nasrallah, 1993).

Under high-temperature conditions, WT plants produced 20.4±4.4 seeds per silique (Fig. 2B), indicating that high temperature adversely affected fertility. Seed numbers of the AlSRK_b+AlSCR_b and AlSRK_a+AlSCR_a transformants grown at 29 °C were not significantly different from those of the WT plants. This confirmed the high-temperature defect in self-incompatibility of the AlSRK_b+AlSCR_b and AlSRK_a+AlSCR_a transformants that was observed in the pollination assays. Two of three AlSRK₃₉+AlSCR₃₉ transformants grown at 29 °C produced almost no seeds per silique, which was consistent with the pollination assay results and showed that the AlSRK₃₉+AlSCR₃₉ transformants retained the self-incompatibility phenotype at 29 °C (Fig. 2A).

Reduced levels of plasma membrane-localized AlSRKb but not AlSRK39 under high-temperature conditions

Since high temperature affected the stigmatic self-incompatibility response in AlSRK_b-FLAG+AlSCR_b transformants (Fig. 1C), we examined the accumulation and subcellular localization of AlSRK proteins under normal (23 °C) and high-temperature (29 °C) conditions. Transformants were developed that expressed AlSCR39 under the control of the AlSCR_b promoter and AlSRK39-FLAG (AlSRK39 tagged at the C-terminus with three tandem FLAG tags) under the control of the AtS1 promoter. Six of eleven AlSRK₃₉-FLAG+AlSCR₃₉ transformants inhibited self-pollen germination and pollen tube growth at 23 °C (Table 1 and Fig. 3A). Two independent AlSRK₃₉-FLAG+AlSCR₃₉ transformants showed an intense self-incompatibility response under high-temperature conditions. These results showed that the C-terminal FLAG tag did not affect AlSRK39 function.

Fig. 3.

Determination of subcellular localization of AlSRKb and AlSRK39 in stigmas of Arabidopsis thaliana at high temperature. (A) Self-pollinated AlSRK₃₉-FLAG+AlSCR₃₉A. thaliana transformants at normal temperature (23 °C, left panel) or treated at 29 °C for 16 h before pollination (right panel). Plants expressing C-terminus-FLAG-fused-AlSRK39 inhibited pollen germination, indicating that the C-terminus FLAG fusion did not affect AlSRK39 function. Scale bar is 100 µm. (B) Total proteins from flower buds of wild-type (WT) plants, and transformants expressing AlSRKb-FLAG and AlSRK39-FLAG at 23 °C, and transformants treated at 29 °C for 16 h prior to bud collection were analysed by immunoblotting (IB) with anti-FLAG antibody (upper panel). The membrane used for immunoblot analysis was stained with Coomassie Brilliant Blue (CBB) as a loading control (lower panel). (C) Total flower-bud proteins were treated (+) with Endo H at 37 °C for 1 h; untreated proteins are indicated by ‘–’. The upper panel shows (IB analysis with anti-FLAG antibody. Glycosylated (arrowhead) and deglycosylated (arrow) forms of AlSRKb-FLAG and AlSRK39-FLAG are indicated. The lower panel shows the membrane stained with CBB as a loading control. The graph shows the percentage of AlSRKb-FLAG and AlSRK39-FLAG in the Endo-H resistant form, which indicates that AlSRK proteins entered the Golgi apparatus (see text for details). Data are means (±SE) of three replicates: dots indicate the value of each replicate. Different letters indicate significant differences between means as determined using the Tukey–Kramer method (P<0.05). (This figure is available in colour at JXB online.)

Total proteins prepared from the buds of AlSRK_b-FLAG+AlSCR_b and AlSRK₃₉-FLAG+AlSCR₃₉ transformants grown at 23 °C and 29 °C were analysed by immunoblotting with anti-FLAG antibody. The accumulation of the AlSRKb-FLAG and AlSRK39-FLAG proteins was comparable in transgenic plants grown at 23 °C and at 29 °C (Fig. 3B), indicating that the elevated temperature did not affect their accumulation in the stigmas.

We then used endoglycosidase H (Endo H) treatment to examine the subcellular localization of AlSRKb-FLAG and AlSRK39-FLAG. Glycans in the endoplasmic reticulum (ER) are high in mannose and Endo H-sensitive, whereas those in the Golgi apparatus are further modified and become Endo H-resistant. Endo H treatment can thus be used to distinguish plasma membrane-targeted glycoproteins retained in the ER from those that entered the Golgi apparatus before transport to the cell surface, and it has previously been used to monitor the plasma membrane localization of AlSRKb (Yamamoto et al., 2014). Consistent with this previous study, 52±5.3% of AlSRKb-FLAG was Endo H-resistant in plants in the normal temperature treatment, as demonstrated by its slower migration relative to the Endo H-sensitive form during SDS-PAGE analysis (Fig. 3C). Most of the AlSRK39-FLAG (85±5.7%) occurred in an Endo H-resistant form. The proportion of Endo H-resistant AlSRKb-FLAG was substantially lower at 29 °C than at 23 °C (11±0.9% at high temperature versus 52±5.3% at normal temperature), indicating that the high temperature affected the transport of AlSRKb-FLAG to the plasma membrane. Conversely, a large proportion (75±1.1%) of AlSRK39-FLAG was Endo H-resistant at high temperature, suggesting that elevated temperature did not affect the transport of AlSRK39-FLAG to the plasma membrane, consistent with the results of the pollination assays.

Conclusions

High temperature affected the stigmatic but not the pollen self-incompatibility response in AlSRK_b-FLAG+AlSCR_bA. thaliana transformants. The levels of plasma membrane-localized AlSRKb-FLAG were significantly lower at elevated temperature (29 °C) than at normal temperature (23 °C). Conversely, AlSRK₃₉+AlSCR₃₉ transformants retained the self-incompatibility phenotype at high temperature and the plasma membrane localization of AlSRK39-FLAG was not affected. These results suggest that the breakdown in the self-incompatibility response induced by high temperature is caused by a defect in AlSRK transport to the plasma membrane. Since proteins are transported to the plasma membrane after forming their tertiary structures in the ER, we speculate that AlSRK39 is able to form its tertiary structure more effectively than AlSRKb, and that the efficiency of formation may reflect the differences observed in the high-temperature tolerance of the self-incompatibility function between S haplotypes. Identification of S haplotypes that produce stable SRK proteins may be useful for plant breeding in the coming decades, over which period temperatures are expected to increase as a consequence of climate change. Stable production of SRK proteins at elevated temperatures would facilitate the production of parental lines that exhibit consistent self-incompatibility phenotypes for F₁ seed production in vegetable Brassicaceae.

Supplementary data

Supplementary data are available at JXB online.

Table S1. PCR primers used for plasmid construction.